An EM Algorithm for Mixtures of Hyperspheres - Computational Imaging and Vision Access content directly
Conference Papers Year : 2022

An EM Algorithm for Mixtures of Hyperspheres


This paper studies a new expectation maximization (EM) algorithm to estimate the centers and radii of multiple hyperspheres. The proposed method introduces latent variables indicating to which hypersphere each vector from the dataset belongs to, in addition to random latent vectors having an a priori von Mises-Fisher distribution characterizing the location of each vector on the different hyperspheres. This statistical model allows a complete data likelihood to be derived, whose expected value conditioned on the observed data has a known distribution. This property leads to a simple and efficient EM algorithm whose performance is evaluated for the estimation of hypersphere mixtures yielding promising results.
Fichier principal
Vignette du fichier
revision_eusipco_2022v1.pdf (679.98 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03693951 , version 1 (13-06-2022)


  • HAL Id : hal-03693951 , version 1


Julien Lesouple, Philippe Burger, Jean-Yves Tourneret. An EM Algorithm for Mixtures of Hyperspheres. 30th European Signal Processing Conference (EUSIPCO 2022), European Association for Signal Processing (EURASIP), Aug 2022, Belgrade, Serbia. ⟨hal-03693951⟩
107 View
110 Download


Gmail Facebook X LinkedIn More